Natural Language Queries

GoPie's natural language interface transforms the way you interact with data. Instead of writing SQL, simply ask questions in plain English, and GoPie's AI will understand your intent and generate the appropriate queries.

How It Works

GoPie uses advanced language models to:

1. Understand Intent - Parse your question to identify what you want to know
2. Map to Schema - Match your terms to actual tables and columns
3. Generate SQL - Create optimized queries for your database
4. Execute & Format - Run queries and present results clearly
5. Suggest Visualizations - Recommend appropriate charts for your data

Getting Started with Natural Language

Basic Questions

Start with simple questions about your data:

"Show me total sales by month"
"What are our top 10 products?"
"Which regions have the highest revenue?"
"Compare this year's sales to last year"
"Find customers who haven't ordered in 90 days"

-- "Show me total sales by month"
SELECT 
  DATE_TRUNC('month', order_date) as month,
  SUM(total_amount) as total_sales
FROM sales
GROUP BY month
ORDER BY month DESC;

"How many new users signed up this week?"
"What's the average session duration?"
"Show user growth over the last 6 months"
"Which features are most popular?"
"Find power users (top 10% by activity)"

-- "How many new users signed up this week?"
SELECT COUNT(*) as new_users
FROM users
WHERE created_at >= CURRENT_DATE - INTERVAL '7 days';

"What's our current burn rate?"
"Show revenue by customer segment"
"Calculate gross margin by product"
"What are our biggest expenses?"
"Project revenue for next quarter"

-- "Calculate gross margin by product"
SELECT 
  product_name,
  SUM(revenue) as total_revenue,
  SUM(cost) as total_cost,
  (SUM(revenue) - SUM(cost)) / SUM(revenue) * 100 as gross_margin_pct
FROM financial_data
GROUP BY product_name
ORDER BY gross_margin_pct DESC;

Best Practices for Natural Language Queries

1. Be Specific

2. Include Time Frames

Always specify when you want data from:

• "last 30 days"
• "this quarter"
• "between January and March 2024"
• "year to date"
• "compared to same period last year"

3. Specify Aggregations

Be clear about how you want data grouped:

• "by month"
• "per customer"
• "grouped by region"
• "daily average"
• "running total"

4. Use Business Terms

GoPie learns your terminology:

• If you call revenue "bookings", use that term consistently
• Define customer segments by your business rules
• Use your product names and categories

Advanced Query Patterns

Comparisons and Trends

"Compare this month's revenue to last month"
"Show the trend of user signups over the last year"
"What's the growth rate of active users?"
"How do sales this quarter compare to our target?"

Filtering and Conditions

"Show only premium customers in California"
"Find orders over $1000 placed on weekends"
"List products with inventory below reorder point"
"Customers who bought X but not Y"

Complex Calculations

"Calculate customer lifetime value by cohort"
"What's the conversion rate from trial to paid?"
"Show moving 7-day average of daily revenue"
"Calculate year-over-year growth by product line"

Statistical Analysis

"What's the standard deviation of order values?"
"Show the distribution of customer ages"
"Find outliers in transaction amounts"
"Calculate correlation between price and sales volume"

Context-Aware Conversations

GoPie maintains context throughout your conversation:

Follow-up Questions

After getting initial results, you can:

• "Filter that to only show values over $10,000"
• "Add a column for percentage of total"
• "Now show me the same thing for last year"
• "Export this to CSV"
• "Create a chart from this data"

Clarifying Ambiguity

When GoPie isn't sure, it will ask for clarification:

You: "Show me user activity"

GoPie: "I found several activity-related columns. Which would you like to see?

• Last login date
• Total sessions
• Actions performed
• Time spent"

Query Suggestions and Auto-Complete

As you type, GoPie provides:

Smart Suggestions

• Common questions for your dataset
• Frequently used filters
• Popular time ranges
• Relevant metrics

Column Recognition

GoPie automatically recognizes:

• Date/time columns for time-based analysis
• Numeric columns for calculations
• Category columns for grouping
• ID columns for joining

Template Questions

Use question templates for consistency:

"Show me [metric] by [dimension] for [time period]"
"Compare [metric] between [segment A] and [segment B]"
"What's the [calculation] of [metric] grouped by [dimension]"

Handling Special Cases

Null Values

Be explicit about how to handle missing data:

• "Include records with missing values"
• "Exclude null prices"
• "Treat missing as zero"
• "Show only complete records"

Date Handling

Specify date formats and timezones:

• "Convert all times to PST"
• "Use fiscal year starting in July"
• "Show dates in MM/DD/YYYY format"
• "Group by week starting Monday"

Currency and Units

Clarify units when needed:

• "Show revenue in thousands"
• "Convert prices to EUR"
• "Display weights in metric"
• "Round to nearest hundred"

Performance Tips

1. Start Broad, Then Narrow

Good approach:
1. "Show me all sales data" (see structure)
2. "Now just show last month" (filter)
3. "Group by product category" (aggregate)

2. Use Sampling for Large Datasets

"Show me a sample of 1000 transactions"
"Use 10% sample for initial analysis"
"Preview first 100 rows"

3. Leverage Indexes

Ask about commonly filtered columns:

• Dates (usually indexed)
• IDs (primary keys)
• Status fields
• Category columns

Common Pitfalls and Solutions

Ambiguous Column Names

Problem: "Show revenue" when you have revenue, net_revenue, gross_revenue

Solution: Be specific: "Show gross revenue before discounts"

Time Zone Confusion

Problem: "Sales yesterday" might mean different things

Solution: Specify timezone: "Sales yesterday in PST"

Aggregation Errors

Problem: "Average order value" without specifying grouping

Solution: Clear grouping: "Average order value by month"

Integration with Other Features

From Natural Language to SQL

After GoPie generates a query, you can:

1. View the generated SQL
2. Edit it in SQL Playground
3. Save it for reuse
4. Share with team members

Creating Visualizations

Natural language works for charts too:

• "Create a line chart of monthly revenue"
• "Show this as a pie chart"
• "Make a bar chart comparing regions"
• "Plot sales over time with a trend line"

Exporting Results

Request exports directly:

• "Export this to Excel"
• "Send me a CSV of these results"
• "Create a PDF report"
• "Generate an API endpoint for this query"

Tips for Success

1. Start Simple - Begin with basic questions and add complexity
2. Use Examples - Provide sample values: "like 'Electronics' or 'Clothing'"
3. Iterate - Refine your questions based on results
4. Save Favorites - Save commonly used queries for quick access
5. Learn from SQL - Review generated SQL to understand the logic

What's Next?

• SQL Playground - Write and edit SQL directly
• Creating Charts - Visualize your query results
• Query History - Manage and reuse queries